Obesity is a major, growing public and personal health problem in the US. The recent increase in obesity has not been generated by any genetic change in human populations but in changes in environment, primarily the availability and consumption of relatively high calorie foods. However, much of the variation in levels of obesity has a heritable basis. This seeming contradiction is resolved through the realization that there is substantial genetic variation in human populations for response to obesogenic environments, such as consumption of a high fat diet. We study the relationship between genetic variation and dietary responsiveness in a mouse model system, the intercross of LG/J and SM/j inbred mouse strains. This cross segregates for many small effect genes that interact with each other, sex, and environment. Our quantitative trait locus (QTL) mapping studies in the F16 generation have identified 40 genomic locations affecting fat depot size. These effects are most often sex and diet-specific and have been mapped to fine genomic resolution (2 - 4 Megabases). We propose to further limit QTL support intervals to 0.50 - 1 Mb using data already in hand on an additional Advanced Intercross (AI) Line generation (F34) and by combining the F16 and F34 data in a joint analysis. The 5-10 positional candidate genes for each QTL will be evaluated with regard to gene expression in male and female LG/J and SM/J animals reared on a low or high fat diet. Sequence polymorphisms, including the 4.1 million detected in our LG/J and SM/J sequencing project, will be examined bioinformatically for potentially functional SNPs in coding, regulatory, and intergenic DNA. As many QTL showed patterns consistent with genomic imprinting, we will also compare methylation and allele-specific expression in hybrids between our parental strains. Sequence, expression, and methylation results will be used to identify the most likely of the positional candidate genes as the gene responsible for QTL effects. This hypothesis will be tested using a quantitative hybrid complementation test where an AI Line heterozygote for the locus in question is crossed with a strain in which the gene of interest is knocked-out. A significant finding that the effect of the KO depends on whether it is mated to a LG/J or SM/J strain shows a failure to complement and demonstrates that the experimental stains carry phenotypically and functionally distinct alleles at the gene in question. We plan to test at least 10 positional candidate genes with this test over the project period. Successful experimental results will identify the genes that cause variation in obesity and in an obese response to a high fat diet. The physiology of these genes and characterization of the molecular polymorphism responsible for the difference will be the subject of future research.